Assign work to AI agents. Keep humans in control.

An AI agent is software that can read context, use tools, and carry out multiple steps toward a goal—not just return a chat response. In Truetask, an agent can be represented as a teammate, assigned a task, and connected to the workspace through MCP.

No. Connect the MCP-compatible agent or client your team already uses. Truetask supplies project context, governed actions, shared progress, questions, and approval. Managed Truetask AI handles in-product assists, while Enterprise adds agent-loop nodes (multi-step AI nodes inside automations).

Create a scoped MCP credential, connect it to a compatible client, and assign the agent a prepared task. The agent can read task context, start or update a structured run, log progress and evidence, ask questions, and act through the tools it is allowed to use.

You decide. Agents can propose completion with evidence, while a human approves or declines. Per-board rules can deny agents the ability to complete, archive, delete, or merge pull requests. Workflows and personas (per-agent limits on tools, iterations, and budget) can also restrict write access, token spend, and time.

The run can enter Needs input. The question appears on the card and in the Agent command center. A human can answer with text or a provided choice, and the agent reads the response on its next cycle.

Yes. It is a full project-management platform for software teams, with nine board views, dependencies, milestones, saved views, notes, Git workflows, forms, automations, time tracking, roles, and realtime collaboration.

GitHub, GitLab, and Gitea, including cloud and self-managed provider instances. Truetask can link tasks to branches, commits, issues, and pull requests, show CI state, and support code review inside the product.

Yes. Bring-your-own-agent access through MCP is available on every plan. Exact credential limits and plan allowances are displayed from the live pricing catalog.

Each cloud workspace runs as its own isolated instance with its own database and scoped secrets. AI keys are stored server-side rather than exposed to the browser. Higher plans add further identity, audit, and retention controls.

Yes. Export your entire workspace as JSON at any time, on every plan. Notes can also be exported as Markdown and list data as CSV, so your team always has a readable copy of its work.

MCP (Model Context Protocol) is an open standard that lets AI tools such as Claude Code use another product's capabilities as callable tools. Truetask exposes 258 MCP tools covering boards, tasks, notes, Git, automations, and webhooks, and MCP access is available on every plan. Setup details live at truetask.io/developers.

Truetask does not use your workspace data to train machine-learning models. AI features call the provider you configure: your own provider keys, or managed Truetask AI, where our agreements with AI infrastructure providers do not permit training on your content. Bring-your-own-key traffic runs under your agreement with that provider. Provider keys are stored server-side, never exposed to browsers. More at truetask.io/trust.

Yes. Truetask ships importers for Trello, Asana, Jira, Notion, Linear, GitHub Projects, and Todoist, and the import dialog shows exactly what maps before anything is written. Step-by-step guides live at truetask.io/switch/trello and the matching pages for each tool.